Abstract:To efficiently evaluate the influence of the distribution parameters of the input variables on the failure probability of engineering structures and improve the reliability and safety of engineering structures in a targeted manner, new methods for the global reliability sensitivity analysis (RSA) of distribution parameters are proposed in this study based on the cubature formula (CF), surrogate sampling probability density function (SSPDF), and quasi-Monte Carlo (QMC) method. By introducing CF, the proposed methods can effectively improve the computational efficiency of the nested expectation and variance operators in the reliability sensitivity indices of the distribution parameters. Based on the concept of SSPDF, a surrogate importance sampling probability density function was developed. This not only overcomes the problem of the computational effort of propagating parameter uncertainty to the failure probability function (FPF), which depends on the dimensionality of the parameters; it also further improves the efficiency of the RSA of the parameters in the case of a small failure probability. Finally, by incorporating the idea of the QMC method, the process of calculating the reliability sensitivity indices of the parameters is reduced from a double-loop to a single-loop one. Three engineering examples are used in this study to demonstrate the efficiency and accuracy of the new algorithms.

Conditional expectation method for estimating parameter sensitivity of unreliability

Estimation of Dependability Measures and Parameter Sensitivities of a Consecutive- -out-of- :F Repairable System with -Step

Estimation of Dependability Measures and Parameter Sensitivities of a Consecutive- $k$ -out-of- $n$ : F Repairable System with $(K-1)$ -Step Markov Dependence by Simulation

Conditional Monte Carlo Estimation of Quantile Sensitivities

Simultaneous estimation of complementary moment independent sensitivity measures for reliability analysis

Estimation of dependability measures of Large-scale Markov-dependent consecutive-k-out-of-n: F repairable systems

An efficient method based on Bayes’ theorem to estimate the failure-probability-based sensitivity measure

Sensitivity estimation of conditional value at risk using randomized quasi-Monte Carlo

A new efficient simulation method based on Bayes' theorem and importance sampling Markov chain simulation to estimate the failure-probability-based global sensitivity measure

Estimating Dependability Measurements Of Consecutive-K-Out-Of-N : F Repairable System With Non-Exponential Repair Time Distribution And (K-1)-Step Markov Dependence By Fast Simulation

Efficient Methods for Reliability Sensitivity Analysis of Distribution Parameters and Their Application

Dependability Estimation for Non-Markov Consecutive-K-Out-Of-N: F Repairable Systems by Fast Simulation

Estimating Sensitivities of Portfolio Credit Risk Using Monte Carlo

Pathwise Estimation of Probability Sensitivities Through Terminating or Steady-State Simulations

Simulating Sensitivities of Conditional Value at Risk

Efficient metamodel-based importance sampling coupled with single-loop estimation method for parameter global reliability sensitivity analysis

Efficient Sensitivity Analysis for Parametric Robust Markov Chains

Two Efficient AK-Based Global Reliability Sensitivity Methods by Elaborative Combination of Bayes’ Theorem and the Law of Total Expectation in the Successive Intervals Without Overlapping

Modelling and estimation of system reliability under dynamic operating environments and lifetime ordering constraints

Efficient variance-based reliability sensitivity analysis for Monte Carlo methods

A New Method for the Fast Simulation of Models of Highly Dependable Markov System